MusGU+ Evaluation: JAM

The model requires a CUDA-compatible GPU with sufficient VRAM (+8GB is recommended). Not feasible to train the model on CPU.

No clear information about the dataset size is provided. The model was trained on a dataset of ~54k hours of audio, which makes adaptation infeasible with musician’s own data.

JAM provides complete code and documentation for pretraining, supervised fine-tuning, and direct preference optimization. Model checkpoints are available.

Adaptation requires technical expertise. All training a preprocessing rely on command-line implementation. No user-friendly adaption interface is provided.

The use, modification and distribution of JAM is subject to Stability AI Community License Agreement, which is restricted to research and non-commercial use, and requires attribution and notice. Commercial-use may be obtained by registering with Stability AI or with a separate commercial license.

There is a hosted HuggingFace web demo, but it is currently unavailable. However, the Gradio code is accessible and can be run locally, though it requires installation and environment setup by the user.

JAM’s source code is openly available and can be freely used without limits, paywalls, or subscriptions.

No clear information about the time required for generation is provided. No reference to real-time generation is provided in the documentation or the research article.

There are no clear instructions or specific tools to integrate JAM to a musician’s workflow.

Outputs generated with JAM cannot be used for commercial-use, and must not be including content that violates copyright laws. Responsibility of the generated outputs relies entirely with the end user.

Support exists via GitHub issues and documentation, but there is no dedicated musician-oriented support space.

JAM can be conditioned by lyrics (including word- and phoneme-level timing) and style (via text or audio).

JAM supports precise time-varying control over vocal structure, with explicit word- and phoneme-level timing and duration inputs that directly guide generation at the latent space. However, no time-varying control is provided for the accompaniment or instrumental structure.

JAM separates lyrics timing, global duration, and style conditioning through explicit control pathways, enabling partial disentanglement of vocal structure. However, musical attributes are not independently controllable, and style remains an entangled representation, limiting fully interpretable, attribute-specific control.

JAM provides several generation control parameters, including global duration, separate guidance strengths for lyrics and style, and diffusion step configuration, allowing users to directly influence generation behavior. However, direct manipulation of internal representations is not supported.